Become a partner

Fighting hoaxes and hate speech on social networks using AI

/ Lecture

When comments on social networks are dominated by a minority with extremist views, it indirectly affects the opinions of the "silent majority" and the extreme suddenly becomes the new normal. Comment moderation is not the only solution to society's polarization, but it can significantly contribute to the solution. However, moderating all comments, refuting every hoax is not practically feasible due to the scope. Artificial intelligence that could recognize suspicious comments (also in the Slovak language) could be a solution. However, the appropriate use of AI does not consist in creating the ultimate censor, but in ensuring that the most harmful posts are automatically identified and that the moderator has a tool at his disposal for the most effective response. This is exactly what we tried to do within our project, while in the meantime we have already managed to achieve the first results in practice.


Juraj Bezděk

Blip solutions
Ing. Juraj Bezděk, for many years devoted himself to research and applications of artificial intelligence in the industrial environment and on the electricity market. He graduated from the Technical University in Bratislava, Faculty of Electrical Engineering and Computer Science. He devoted himself to several applications of the use of AI, such as fault analysis in the Slovak language, predictive maintenance, spatial computer vision, or forecasts of consumption and supply of electricity on the electricity market. He worked on important projects for customers such as Slovnaft, Duslo, ZSSK and the Slovak Electricity Transmission System. At Blip solutions, he collaborates on tools to democratize and simplify access to AI based on text analy…
See more info about the speaker
Páčil sa ti článok? Zdieľaj ho a povedz o ňom aj ostatným
Nastavenie súborov cookies